Autonomous Driving Validation Using Cloud Comparison and Fleet Diagnostics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving systems lack comprehensive validation and predictive maintenance capabilities, particularly in identifying sensor malfunctions, miscalibration, and vehicle performance issues, which can lead to safety concerns and operational inefficiencies.
Innovation Solution
A large-scale autonomous driving validation system that integrates on-board primary advanced autonomy systems with off-board high-performance cloud computing, enabling real-time comparison of computations and sensor data to detect potential problems, generate notifications, and facilitate remote supervision and predictive maintenance across fleets of vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If on-board primary autonomy systems process computations in real-time, then response speed is improved, but measurement precision and reliability of detection deteriorate due to limited computational resources
Solution Approach 1:
The system divides computational tasks between on-board primary autonomy systems (for real-time emergency responses) and off-board cloud computing systems (for comprehensive analysis and validation). This segmentation allows each subsystem to operate within its computational capacity while achieving overall system precision through distributed processing.
Solution Approach 2:
An off-board cloud computing system acts as an intermediary that receives sensor data and computation results from multiple autonomous vehicles, performs validation computations, and provides feedback. This intermediary enables enhanced detection precision without compromising real-time response capabilities of individual vehicles.
2Reliability
If comprehensive sensor validation and predictive maintenance are implemented, then reliability is improved, but device complexity increases due to additional validation systems
Solution Approach 1:
The off-board cloud computing system serves multiple functions simultaneously: it validates sensor data from multiple vehicles, performs predictive maintenance analysis, conducts computation validation, and provides fleet-wide insights. This multi-functionality achieves comprehensive reliability improvement without proportionally increasing overall system complexity.
Solution Approach 2:
The system implements self-validation through automated comparison of computations between primary and validation systems, and uses machine learning processes to automatically identify sensor malfunctions, miscalibration, and component failures without requiring manual intervention for each validation task.
3Measurement precision
If multiple autonomous vehicles share data and computations are validated across the fleet, then measurement precision is improved, but loss of time increases due to data transmission and processing delays
Solution Approach 1:
The system performs validation computations in parallel with primary autonomy computations rather than sequentially. The off-board system receives and processes sensor data simultaneously as vehicles operate, conducting validation analyses without requiring vehicles to wait for validation results before continuing operations.
Solution Approach 2:
The validation system operates continuously in the background, constantly receiving and analyzing sensor data from the fleet without interrupting vehicle operations. This continuous validation process maintains measurement precision while minimizing time loss through uninterrupted parallel processing.
4Reliability
If real-time computation validation is performed across the fleet, then reliability is improved, but use of energy increases due to continuous data transmission and processing
Solution Approach 1:
The system implements differentiated data transmission where only critical sensor data and computation results are transmitted to the off-board system in real-time, while less critical data is transmitted asynchronously or in batches. This local optimization of data transmission quality reduces energy consumption while maintaining essential validation capabilities.
Data Source
AI summary
An autonomous driving validation system may include a primary advanced autonomy system located on board an autonomous vehicle (AV) that includes sensors, computing subsystems, and a drive-by-wire system. The primary advanced autonomy system may be configured to process first computations relating to a driving operation of the AV. The autonomous driving validation system may also include an off-board autonomous cloud computing system that includes cloud computing subsystems. The off-board autonomous cloud computing system may be configured to process second computations that are similar to the first computations processed by the primary advanced autonomy system. The autonomous driving validation system may include processors and non-transitory computer-readable storage media configured to store instructions that, in response to being executed, cause a system to perform operations relating to whether a likelihood of a problem arising during driving operation of the AV is above a threshold and generating a notification message regarding the problem.